Skip to content

Keep new training model repositories private - #297

Open
voropaevv wants to merge 1 commit into
huggingface:mainfrom
voropaevv:codex/private-training-repos
Open

voropaevv wants to merge 1 commit into
huggingface:mainfrom
voropaevv:codex/private-training-repos

Conversation

@voropaevv

Copy link
Copy Markdown

Fixes #287.

The SFT, DPO, GRPO, and Unsloth examples now request private repositories when creating their Hub destinations. Both final Unsloth upload paths pass the same visibility setting.

A local stub check captured the arguments for all six paths. The generated manifests, version bump, skill index, and distribution checks passed.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

huggingface-llm-trainer: example training scripts push models to new public repos by default

1 participant